It's useful to know how the .po-file has to look like when using ngettext:
msgid "item"
msgid_plural "items"
msgstr[0] "Produkt"
msgstr[1] "Produkte"
In php:
echo ngettext('item', 'items', $number);
(PHP 4 >= 4.2.0, PHP 5, PHP 7, PHP 8)
ngettext — Version plurielle de gettext
Version plurielle de gettext(). Quelques langues ont plus d'une forme de messages pluriels dépendant du compteur.
singular
L'ID singulier du message.
plural
L'ID pluriel du message.
count
Le nombre (i.e. nombre d'éléments) pour déterminer la traduction du nombre grammatical respectif.
Retourne un message au pluriel identifié par
msgid1
et msgid2
pour le compteur n
.
Exemple #1 Exemple avec ngettext()
<?php
setlocale(LC_ALL, 'cs_CZ');
printf(ngettext("%d window", "%d windows", 1), 1); // 1 okno
printf(ngettext("%d window", "%d windows", 2), 2); // 2 okna
printf(ngettext("%d window", "%d windows", 5), 5); // 5 oken
?>
It's useful to know how the .po-file has to look like when using ngettext:
msgid "item"
msgid_plural "items"
msgstr[0] "Produkt"
msgstr[1] "Produkte"
In php:
echo ngettext('item', 'items', $number);
Even though "hek at theeks dot net"'s answer is valid, I would not recommend using the abs() hack recommended. Even though it is by far the most common, not all languages treat (n != 1) as plural. Other languages are much more complex, for example, here is how you determine plurals in Macedonian.
n==1 || n%10==1 ? 0 : 1
In Arabic there are actually 5 different types of plurals:
n==0 ? 0 : n==1 ? 1 : n==2 ? 2 : n%100>=3 && n%100<=10 ? 3 : n%100>=11 ? 4 : 5
If you are using only specific languages that use the (n != 1) format AND -1 is singular, by all means, use abs(), but be careful and don't forget that you have done this when adding a new language to your project 3 years down the road.
According to GNU gettext manual third argument is unsigned long integer. It must be positive number. If n is negative, it might be evaluated incorrectly in some languages.
Example for russian lang:
file.po:
...
"Plural-Forms: nplurals=3; plural=n%10==1 && n%100!=11 ? 0 : n%10>=2 && n%10<=4 && (n%100<10 || n%100>=20) ? 1 : 2;\n"
...
msgid "File"
msgid_plural "Files"
msgstr[0] "Файл"
msgstr[1] "Файла"
msgstr[2] "Файлов"
...
file.php
...
echo ngettext("File", "Files", $number);
...
Beware of one difference between the GNU gettext API and the PHP binding of it, which is that the GNU gettext functions that accept a $count parameter all expect (indeed, being compiled C, require) that $count be unsigned, while the PHP binding does not.
Thus, the PHP gettext functions will happily accept negative numbers. The one potentially irritating consequence of this is that -1 is treated as plural, which sits well with some people and not so well with others. (As a picky native speaker of English, my personal opinion is that both "the temperature is minus one degree Fahrenheit" and "four apples minus five apples leaves minus one apple" but others may feel that "four apples minus five apples leaves minus one apples" sounds better.)
The upshot: You may want to abs($count) before passing numbers to gettext.
Bonus points: If your application includes user preferences, you might offer a "treat -1 as singular" option to your users, then choose $count or abs($count) to pass to gettext based on each user's preference.
Section 10.2.5 in the GNU gettext manual explains the ngettext function:
http://www.gnu.org/software/gettext/manual/
(Sorry, but the Add Note function prevents me from including a long URL which points right to that section of the manual.)